RT info:eu-repo/semantics/article
T1 An explainable deep learning approach for sleep staging in sleep apnea patients across all age subgroups from pulse oximetry signals
A1 Vaquerizo Villar, Fernando
A1 Gutierrez Tobal, Gonzalo César
A1 Álvarez González, Daniel
A1 Martín Montero, Adrián
A1 Gozal, David
A1 Hornero Sánchez, Roberto
K1 Age subgroups
K1 Deep learning
K1 Explainable artificial intelligence
K1 Pulse oximetry
K1 Obstructive sleep apnea
K1 Sleep stages
K1 32 Ciencias Médicas
AB Deep-learning (DL) approaches have been developed using pulse rate (PR) and blood oxygen saturation (SpO2)recordings from pulse oximetry to streamline sleep staging, particularly for obstructive sleep apnea (OSA) pa-tients. However, lack of interpretability and validation across patients from a wide range of ages (children,adolescents, adults, and elderly OSA individuals) are two major concerns. In this study, a DL model based on theU-Net framework (POxi-SleepNet) was tailored to accurately perform 4-class sleep staging (wake, light sleep,deep sleep, and rapid-eye movement sleep) in OSA patients across all age subgroups using PR and SpO2 signals.An explainable artificial intelligence (XAI) methodology based on semantic segmentation via gradient-weightedclass activation mapping (Seg-Grad-CAM) was also applied to quantitatively interpret the time and frequencycharacteristics of pulse oximetry recordings that influence sleep stage classification. Overnight PR and SpO2signals from 17303 sleep studies from six datasets encompassing children, adolescents, adults, and elderly OSAindividuals were used. POxi-SleepNet showed high performance for sleep staging in the six databases, withaccuracies between 81.5 % and 84.5 % and Cohen’s kappa values from 0.726 to 0.779. It also demonstratedgreater generalizability than previous studies. XAI analysis showed the key contributions of mean and variabilityin PR and SpO2 amplitude, as well as changes in their spectral content across specific frequency bands(0.004–0.020 Hz, 0.020–0.100 Hz, and 0.180–0.400 Hz), for sleep stage classification. These findings indicatethat POxi-SleepNet could effectively automate sleep staging and assist in diagnosing OSA across all age groups inclinical settings.
PB Elsevier
SN 0952-1976
YR 2025
FD 2025
LK https://uvadoc.uva.es/handle/10324/79844
UL https://uvadoc.uva.es/handle/10324/79844
LA eng
NO Engineering Applications of Artificial Intelligence, 2025, vol. 162, p. 112562
NO Producción Científica
DS UVaDOC
RD 19-nov-2025